ASSESSING CONSUMER’S WILLINGNESS TO PAY FOR IMPROVED
DOMESTIC WATER SERVICES IN KELANTAN, MALAYSIA
Mahirah Kamaludin, Khalid Abdul Rahim
Institute Of Agricultural And Food Policy Studies, Putra Infoport, University Putra Malaysia
[email protected]
Alias Radam
Department Of Management and Marketing, University Putra Malaysia
ABSTRACT
This study aims to estimate willingness to pay (WTP) of people in the state of
Kelantan, Malaysia for improvement in their domestic water services. Their water
sector has been a long debate since their poor water services have threatened
people’s daily routine such as water disruption, water leakage, smelly and dirty
water supply. The water services should be managed effectively based on efficient
water pricing since any improvement in services involve high cost in its operation. A
Contingent Valuation Method (CVM) was employed to 552 households in the state
for analysis regarding to the services and Probit model was used to analyse the data
obtained through the survey. The results show that bid price, household income and
household size have statistically significant impact on WTP and they are as expected
in earlier studies. The calculated mean WTP is RM0.60 applied on the first 35m 3 and
it is much higher from current water price. The new water price can be
recommended for any improvement and upgraded services to high standard in
services in the future which is satisfying consumer’s needs.
Keywords: Contingent Valuation method, Probit model, willingness to pay, water
services, Kelantan.
1. INTRODUCTION
High rapid population and urbanization has
increased the demand for water and is
expected to undergo substantial changes in
the future. Water companies should manage
the problem as soon as possible in order to
satisfy consumer’s needs. But, low water
prices can limit water companies’ ability to
enhance their service quality and cover their
capital costs. Water supply failing to meet
demand is putting pressure on consumers.
Water supply problems also have clear
impacts on social stability and economic
productivity, thus solutions in dealing with
water shortages are needed. It is important to
know how much amount that households are
willing to pay in order to help with the
development and upgrading projects in water
services. The World Bank Water Demand
Research Team (1993) points out that it is
important to provide the consumers with
services that what they are really wanted and
how much they are willing to pay.
Problems in water services in Kelantan are a
long standing issue due to its failure in
supplying excellent water services to their
[45]
consumers. Currently, water service in the
state is supplied by the sole provider, Air
Kelantan Sdn. Bhd. (AKSB). The water service
in Kelantan was privatized in October 1995
and before this it was under the Water Supply
Division, Public Works Department. The
situation of the water supply was very bad
and it was not keeping pace with the demand,
causing consumer dissatisfaction until the
industry was privatized. The company
operates in ten districts which are Kota Bharu,
Bachok, Pasir Puteh, Pasir Mas, Tanah Merah,
Jeli, Tumpat, Machang, Kuala Krai and Gua
Musang. Most of Kelantan’s population is
concentrated in one-third of the state area
with the availability of water in its alluvial
subsurface terrain (Zamri, 2009). About 80%
of the population lives in the northern region
of the low alluvial plain and the remaining
20% live in the southern parts on higher
ground. Kelantan’s water sources consist of
60% from surface water and 40% from
groundwater (Zamri, 2009). Presently,
Kelantan is the largest groundwater operator
in Malaysia (The Malaysian Water Association,
2011). Groundwater utilization is significant in
less developed states such as Kelantan, and
Sabah and groundwater is only seen as a
substitute for surface water in certain places
(Lee, 2007). Groundwater will be an essential
source in meeting future water demands for
the public supply.
The objectives of this study are to determine
how much people in Kelantan are willing to
pay (WTP) for improvement in the quality of
domestic water services and what are possible
determinants affecting their WTP. The
organisation of this paper as follows, Section 1
and Section 2 introduces the background of
the study, Section 3 explains the methodology
and its application in the study, Section 4
demonstrates the results and discussions and
finally Section 5 concludes the study.
2. BACKGROUND
Kelantan is one of the states in Malaysia,
whose capital and royal seat is situated in Kota
Bharu. Kelantan is the most Northeastern
state of Peninsular Malaysia and it is bordered
by Narathiwat Province of Thailand to the
north, Terengganu to the south-east, Perak to
the west, and Pahang to the south. To the
north-east of Kelantan is the South China Sea.
The state is flourishing with verdant paddy
fields, rustic fishing villages and casuarinas
lined beaches. Kelantan is an agrarian
economy and the state is driven by the
production of rubber, rice, and tobacco. The
population of Kelantan was about 1.6 million
in 2011 (Department of Statistic, 2012). The
9th Malaysia Plan demonstrates that Kelantan
is ranked at 13th place in Malaysia based on
development composite index by state and a
large share of the population lives in rural
areas. Kelantan is categorized as a state with
high levels of poverty (4.8%) in Malaysia
(Economic Planning Unit, 2009).
A few states which are less developed in
Malaysia struggle to improve water coverage
mostly in rural areas including Kelantan (Lee,
2011). The state demonstrates the lowest
coverage of water supply for both urban
(57.9%) and rural (56.1%) areas by 2010
(Malaysian Water Association, 2011). Kelantan
lacks financial sources in order to improve
their water supply coverage to serve the
public. Most water sources come from
groundwater and groundwater utilization is
most significantly in Kelantan (Malaysian
Water Association, 2011). However, the
groundwater is faced with problems too, such
as over extraction, subsidence and
contamination. A statistic by the Malaysian
Water Association demonstrates water
services complaints based on leakage, water
quality, and water pressure, which shows that
the rising complaint on water quality from
2009 to 2010. Most of people in Kelantan
complaint about the quality of water supply
which is not up to the standard, due to its
[46]
condition, smelly and coloured water supply.
A research by the Association of Water and
Energy Research Malaysia (2011) points out
some cases in Kelantan with dirty and smelly
water supply, low coverage performance and
frequent water disruptions. Presently, water
prices seem too cheap and it is unable to
generate enough revenue to cover the full
cost of capital investment, operation and
maintenance. Kelantan ranked at 3rd place in
2011 which demonstrates lowest water prices
for domestic water services (RM0.55 for first
35m3). If the price is too low, the water
company is unable to maintain and sustain its
operations and if the price of water is too
cheap it leads to water wastage. Low water
price and its sufficient supply makes
consumers take water for granted in water
consumption.
2.1 Consumer’s Willingness to Pay
WTP describes the maximum amount
consumer’s are willing to give up for certain
benefits in exchange for others, rather than
without having it. WTP presents the value of a
good to an individual as what they are willing
to pay, sacrifice or exchange for it. Based on
law of demand, the lower the price of a good,
the larger the quantity consumers wish to pay
(Browning and Zupan, 2004). There are many
previous studies that have existed about WTP
for water services such as in Larson, Lew and
Onozaka (2001), Farolfi, Mabugu and
Ntshingila (2007), Vasquez et al. (2009) and
Bockstael, McConnell and Strand (1989).
Sarala Devi et al. (2009) argue that any water
pricing system should consider factors such as
WTP ability and financial, affordability,
socioeconomic, managerial, technical, and
political factors.
3. METHODOLOGY
The consumer’s maximum willingness to pay
for improvement in water services at Q1, while
the status quo provision of water services is
Q0. When consumers are confronted with a
question either to accept or reject the
improvement program from Q0 to Q1, it is
important to ask consumers regarding their
WTP in order to get the proposed change.
New utility theoretical framework (Y0 – WTP)
shows how the respondents attain satisfaction
when the income reduces by their willingness
to pay. This study exhibits that consumer’s
derived utility can be expressed as in Equation
1:
V (P0, Q0, Y0, Z0)
V (P1, Q1, Y0 - WTP, Z1)
(1)
The consumer’s WTP for water service
improvement is a function of these variables
(provision of water services, income and other
consumer’s characteristics). Equation 2
demonstrates how the changes in income
affecting the WTP. WTP as a dependent
variable which is consumer’s willingness to
pay for a transformation in water services and
it can be expressed in a linear regression form
as below:
WTP = β0 + HINC β1 + WPRC β2 + HSZ β3 + e
(2)
Where;
WTP
=
Probability of saying “Yes” or
“No” to offered prices.
WPRC =
Bid water price expressed in
Ringgit Malaysia (RM).
HINC =
Household’s
income
expressed in Ringgit Malaysia (RM).
HSZ
=
Household size expressed in
number of individual.
3.1 Contingent Valuation Method (CVM)
The CVM can be used to estimate nonmarketed goods such as environmental assets,
amenities and services, and it elicits the
preferences for non-marketed goods (United
Nations Environment Programme, 2007).
Benefits are measured directly rather than
inferred from a demand curve in CVM. In the
CVM, respondents are asked whether or not
[47]
they would pay or accept a specific amount
for a program or policy in question. The
method entails by asking question in this
study such as “Are you willing to pay RM x for
improved in domestic water services?”. The
respondents choose whether to “take it” or
“leave it” to that proposed price. The CVM
uses closed-ended format that can be
answered using a simple "yes" or "no"
regarding to their WTP. It can be presented by
using model formulation which followed
Hanemann et al. (1991) as below:
Prob { No }  Prob { WTPmax <
PRICE }  G (PRICE; θ)
(3)
Prob { Yes }  Prob { WTPmax >
PRICE } 1 - G (PRICE; θ)
(4)
Equation 3 explains that if the proposed price
amount (PRICE) is more than consumer’s
willingness to pay, then they are not willing to
pay for that amount. However, if the bid is
below their true maximum willingness to pay
amount, the probability of answering to that
amount is “yes” based on Equation 4, it
implies they maximize utility and willing to pay
for that specific amount.
3.2 Probit Model
This study has dependent variable which
elicits “yes” or “no” responses to WTP
question, so probit regression are suitable to
estimate the model. The Probit model is based
on normal distribution as below (Gujarati,
2003):
−𝑠2
Prob (Y) =
𝑧
1
∫ 𝑒 2
√2𝜋 −∞
ds
(5)
Assuming Y is a binary which has possible
outcome recoded as 1 or 0, based on
respondent answering yes/no to the
questions. Thus, particularly we assume the
model as:
Prob (Y = 1|𝑋𝑖 ) = Φ (Xi β)
(6)
Where Prob represents probability of
answering to the question, Φ is cumulative
distribution function (CDF), and β can be
estimated by maximum likelihood. Y*i signifies
the value of one if respondents are willing to
pay for improvement, otherwise the value of
zero, if they are against the program.
Y*i = {
1 𝑖𝑓 𝑤𝑖𝑙𝑙𝑖𝑛𝑔 𝑡𝑜 𝑝𝑎𝑦, 𝑌𝑖∗ > 0,
0 𝑖𝑓 𝑛𝑜𝑡,
𝑌𝑖∗ ≤ 0,
(7)
Equivalently, we prove that both models in
Equation 6 are corresponding as follows:
Prob (Y = 1|𝑋𝑖 )
ei > 0)
= Prob (Y* > 0) = Prob (Xi β +
(8)
= Prob (ei > - Xi β)
= Prob (ei < Xi β)
= Φ (Xi β)
Estimated parameters can be achieved by
maximizing the following log likelihood
procedure:
log L (β) = ∑𝑛𝑖=1{𝑦𝑖 𝑙𝑛 𝛷 (𝑥𝑖 𝛽) + (1- 𝑦𝑖 ) ln (1- Φ
(𝑥𝑖 𝛽)) }
(9)
Then, the calculated mean WTP can be
derived as stated in below equation
(Cameron, 1988):
n
WTP 
 0   i X
i-2
- 1
(10)
Where β0 is estimated constant, β1 is the
coefficient for the proposed price amount,
and βi is the coefficient for socio-economic
characteristics of respondents.
3.3 Sampling
There were 552 of domestic water services
users who have registered active accounts
with AKSB as respondents in this study. A
sample survey was conducted for domestic
users which come from urban and rural areas
in Kelantan and they were from 10 of districts
[48]
in Kelantan. This study uses stratified random
sampling based on the districts in the state as
shown in Table 1.
Table 1: Total of Respondents According to
the Districts
No. of
District
Respondents
Kota Bharu
168
Pasir Mas
70
Tumpat
55
Bachok
46
Pasir Puteh
44
4. RESULTS AND DISCUSSION
Tanah Merah
Kuala Krai
Gua Musang
Machang
Jeli
Total
45
38
34
35
17
552
The distributions of the respondents were not
equivalent in each district since the districts
did not have the same total numbers of
population. The respondents were selected
randomly in each stratum.
These findings in Table 2 present that the average household size was 5 with standard deviation of
2.59. The average age of respondents were 38 years and most of them were female (50.2%).
Household income was RM 4077.90 per month in average.
Table 2: Socio Demographic Profile of Respondents
Frequency
Percent (%)
Mean
Characteristics
Gender
Male
Female
Age
20 – 30 years
31 – 40 years
41 – 50 years
51 – 60 years
61 – 70 years
>71 years
Race
Malay
Chinese
Indian
Others
Household Size
1 – 5 people
6 – 10 people
>10 people
Education Level
PhD/Master
Bachelor
Diploma/Certificate
Secondary level
Primary level
No Education
Numbers of working
275
277
49.8%
50.2%
27.7
30.8
29.3
9.6
1.8
0.7
27.7%
30.8%
29.3%
9.6%
1.8%
0.7%
513
32
3
4
92.9%
5.8%
0.5%
0.7%
271
266
15
49.1%
48.2%
2.7%
22
139
200
134
50
7
4%
25.2%
36.2%
24.3%
9.1%
1.3%
[49]
SD
38.34
10.99
5.10
2.59
family members
0 – 3 members
4 – 7 members
>8 members
Household Income
(monthly)
Less than RM 2,000
RM2,001 – RM 4,000
RM4,001 – RM 6,000
RM6,001– RM 8,000
RM8,001– RM 10,000
>RM10,001
481
65
6
87.1%
11.8%
1.1%
162
154
127
51
22
36
29.3%
27.9%
23%
9.2%
4%
6.5%
2.13
1.18
4077.90
2720.711
4.1 Contingent Valuation Method (CVM) Analyses
Respondents answered the proposed water price with “Yes” and “No”, which are coded as “Yes = 1”
and “No = 0”. The respondents are offered five groups of bidding price in this study. Table 3 presents
consumer’s response to five groups of bidding price in the survey according to the districts. The
distribution of price bids and respondents in each districts were not distribute equally since the
population in each districts were not equivalent. Respondents were offered with different water price
bid which were the prices in the range of RM 0.61 (10% increase from current price), RM 0.63 (15%
increase from current price), RM 0.66 (20% increase from current price), RM0.69 (25% increase from
current price), and RM0.72 (30% increase from current price). Table 3 portrays percentage of
respondents rejecting the offered price as the price bid increases. About 70 of respondents reject the
highest price bid (RM0.72) and only 21 of respondents are willing to pay the proposed water price.
Price
Table 3: Willingness to Pay According to the Districts in Kelantan
RM 0.61
RM 0.63
RM 0.66
RM 0.69
RM 0.72
Bid
(per
m3)
District
Kota Bharu
Pasir Mas
Tumpat
Bachok
Pasir Puteh
Tanah Merah
Kuala Krai
Gua Musang
Machang
Jeli
Total
No
Yes
No
Yes
No
Yes
No
Yes
No
Yes
n
12
12
10
9
5
6
3
1
2
1
61
16
2
1
1
3
3
4
6
5
3
44
10
13
10
7
4
7
4
2
2
1
60
33
1
1
2
5
2
4
5
5
3
61
20
12
9
8
4
7
5
2
3
2
72
19
2
2
1
5
2
3
5
4
1
44
17
14
9
9
6
8
6
6
3
1
79
25
0
2
0
3
1
2
1
4
2
40
8
13
11
9
8
5
6
5
4
1
70
8
1
0
0
1
4
1
1
3
2
21
168
70
55
46
44
45
38
34
35
17
105
121
116
4.2 Probit Model Results
[50]
119
91
552
The data is analysed by using LIMDEP, NLogit
Version 9 and value of 0 is given to
respondents who rejected the idea for
improvement, saying “No” to offered price.
Then the value of 1 is for “Yes” answer if they
accept the offer. Based on Table 4,
explanatory variables have a unique
statistically significant contribution to the
model. The ‘Bid Price’ variable has a negative
sign and significant at 1% level as predicted. It
signifies that under hypothetical market, as
offered bid price increases, the probability of
saying “Yes” decreases among consumers.
Respondents reject the offered price as it
increases and this is consistent with demand
theory since consumers react to higher prices.
Household income is a significant determinant
since it illustrates positive sign and significant
at 1% level too. The determinant suits to
economic theory which states that WTP for
“good” increases with income, thus household
income should have positive sign. People tend
to spend more and households with high level
of income are willing to pay higher for water
bill if there is an increase in water prices. The
household size variable shows negative
relationship with WTP but significant at 5%
level. It demonstrates that as household size
increases, the WTP decreases. The more
household members in the house, there will
be a reduction in WTP. It has been
demonstrated that the more households
consume water, the less they are willing to
pay (Farolfi, 2007). They demand more water
consumption but higher water price can be a
burden for them. Thus, it consistent with
demand theory; as price increase, there will
be a decrease in quantity demanded.
Table 4: Final Regression Results of Probit Model
Probit Model
Variables
Coefficient
t-value
Constant
3.8989
3.912***
Bid Price
-6.0315
-4.075***
Household Income
0.0000094
3.927***
Household size
-0.0539
-2.380**
Log Likelihood
-349.7500
Restricted Log Likehood
-366.1891
Chi squared
32.8782***
McFadden
0.0448
R-squared
Percentage Correct
62.14%
Note: (**) 5% level, (***) 1% level.
Regression was performed to elicit
consumer’s WTP towards improvement in
domestic water services and the probit model
has three explanatory variables (bid price,
household income and household size) which
are statistically significant. In probit model,
the chi-squared shows 32.87 and the value
McFadden R squared value is 0.0448. The R
Square expresses an indication of the amount
of variation in the dependent variable clarified
by the model. The model correctly classified
62.14% of cases.
4.3 Estimation of Mean Willingness to Pay
(WTP)
The estimated mean of WTP ranges from
RM0.4637 to RM0.7621 in this study. We
estimate consumer’s willingness to pay for
improved domestic water services in Kelantan
by referring to Equation 10.
WTP = [β0 + (βHINC * HINC + βHSZ * HSIZE)] / (βPRICE)
(11)
[51]
Where, βHINC , βHSZ and βPRICE show estimated
variables for household income, household
size and price bid. Thus, the mean WTP
computed is RM 0.60 for probit model applied
on the first 35 m3. It increases by about 9.09%
from the current water price (RM0.55 applied
on the first 35 m3).
5. CONCLUSION
This study has shown that people in Kelantan
are willing to pay more than the current water
price, which is RM0.60 applied on the first 35
m3 by employing probit model. It
demonstrates the three factors that most
influencing people’s willingness to pay for
improved services based on findings of CVM
approach such as price bid, household income
and household size. The signs indicate as
expected in earlier studies. The high increase
in water demand requires the authorities and
water companies to think of on new strategies
to attain efficient water services. Cheap water
prices
will
limit
implementation
of
infrastructure projects to upgrade facilities
and it implies low value of water, though the
sources are valuable and insufficient. The
government should regulate the market by
controlling water prices in the state. The
prices should be reliable in order to balance
and protect both sides; producers and
consumers. Water prices should reflect the
cost of water production in order to promote
market efficiency. If the water price is
increased, it can lessen the financial burden of
water companies by providing needed funds
for further improvement in infrastructure,
upgrade services and financial development.
Revising prices can help the water companies
to reduce cost somewhat but they still have to
operate until the optimum production is
achieved. The consumers will be aware to
conserve water and avoid water wastage due
to an increase in the water price. Participating
in awareness programs makes the consumers
accountable for water conservation and
appreciates the value of water, the most
valuable among natural resources. Since water
is considered as a ‘public good’, it should be
allocated to the consumers at the best
condition.
REFERENCE
Association of Water and Energy Research
Malaysia. (2011). National water
services industry restructuring –
Kelantan state case study. Kuala
Lumpur.
Bockstael, N.E., McConnell, K.E., and Strand,
I.E. (1989). Measuring the benefits of
improvements in water quality: the
Chesapeake Bay. Marine Resource
Economics. 6: 1-18.
Browning, E.K., and Zupan, M.A. (2004).
Microeconomics:
Theory
and
Applications (8th edition). United
States of America: John Wiley & Sons,
Inc.
Cameron, T.A. (1988). A new paradigm for
valuing non-market goods using
referendum
data:
maximum
likelihood estimation by censored
logistic
regression.
Journal
Environmental
Economics
and
management. 15: 355-379.
Department of Statistic (2012). Kelantan at a
Glance:
Population
in
2011.
http://www.statistics.gov.my/portal/i
ndex.php?option=com_content&id=5
26&lang=en&negeri=Kelantan
Economic Planning Unit. (2009). Incidence of
poverty, by ethnicity, strata and
state,
Malaysia
1970-2009.
Household income and poverty.
http://www.epu.gov.my/c/document
_library/get_file?uuid=5bf3a7ca37a7-4ebb-96c6634ce17141dc&groupId=34492
[52]
Farolfi, S., Mabugu, R. E., and Ntshingila, S. N.
(2007). Domestic water use and
values in Swaziland: a contingent
valuation analysis. Agrekon. 46 (2):
157-169.
Gujarati, D. (2003). Basic Econometrics (4th
edition). New York: Mc GrawHill.
Hanemann,
W.M. (1994). Valuing the
environmental through contingent
valuation. Journal of Economics
Perspectives. 8 (4): 19-43.
Larson, D.M., Lew, D.K., and Onozaka, Y.
(2001). The public’s willingness to pay
for improving California’s water
quality. Paper presented at Annual
Meeting of Western Regional project
W-133. Miami: Florida.
Lee, C.H.K. (2007). Social policies and private
sector participation in water supply –
the case of Malaysia. Retrieved from
United Nations Research Institute for
Social Development, program on
Social Policy, Regulation and Private
Sector Involvement in Water Supply
at: http://www.unrisd.org/
United Nations Environment Programme.
(2007). Guidelines for conducting
economic valuation of coastal
ecosystem goods and services
(UNEP/GEF/SCS Technical Publication
No.8). Thailand: United Nations
Environment Programme.
Vasquez, W.F., Mozumder, P., HernandezArce, J., and Berrens, R.P. (2009).
Willingness to pay for safe drinking
water: evidence from Parral, Mexico.
Journal of Environment Management.
90: 3391-3400.
World Bank Water Demand Research Team
(1993). The demand for water in rural
areas: determinants and policy
implications. The World Bank
Research Observer. 8 (1): 47-70.
Zamri, W. M. (2009). Groundwater for
domestic needs in Kelantan. Water
Malaysia, 20, pp. 6-10.
Lee, C.H.K. (2011). Privatization, water access
and affordability: evidence from
Malaysian household expenditure
data. Economic Modelling, 28, 21212128.
http://www.sciencedirect.com/scienc
e/article/pii/S0264999311001143
Malaysian
Water
Association.
(2011).
Malaysia water industry guide. Kuala
Lumpur: MWA.
Sarala Devi, J., Joseph, B., Karunakaran, K.,
Anurdha, B., and RamaDevi, K.
(2009). People’s attitudes towards
paying for water. Current Science. 97
(9): 1296-1302
[53]